Nice catch! That moment looks weird! I didn’t power the board off at that moment @Pimousse
What is interesting, rpm measurement stays the same, current measurement jumps zero, voltage drops twice and duty cycle increases twice (bit shift?
) Looks like a glitch in VESC!
Oh, then there is something to dig. The 2 points before are 34,8v then 26,2v, both with 0A currents. The duty cycle increases due to PID Control, this is a normal behaviour. All other values are normal as well (temperatures, speed…).
To me, it sounds like you had a micro power cut. But caps maintained the VESC alive. Check your soldering (power connectors and caps).
You also have data transfer to consider. The BLE connection is not super mega fast, update rate is limited a bit and in consequence the readings are not totally accurate. It also depends on the design of the ESC. Two battery side shunt ESCs can’t read currents and voltages as accurate as three high side shunt ESCs with shunt amplifiers and filtering.
I just updated my firmware to the latest version from vesc_tool mobile(1.07). I have cogging(at low speed) just before coming to stop. High speed braking has not cogging. The motor shudders. I am going to run the detection again and see if that solves the problem.
Edit: Went through the same process again, no difference. Still have cogging. Anyone else?
A bit of cogging is normal on the last bit of RPM when braking. This is unavoidable, since at that motor speed tracking of the rotor is very difficult. It depends on many factors. Some motors might cogg less than others and some boards use more gearing and others are poorly geared. Our boards don’t cogg, but you can feel a bit of fine ripple vibration from 2Km/h to zero. This is the speed where the VESC used to short out the motor phases and create max braking strength. If you would use a AS5047P Sensor, you would not feel cogging a all. The halls are just not precise enough and back EMF is not giving good feedback at such low speeds.
That makes sense, but my previous firmware/settings gave me a pretty smooth braking until the very end. So not exactly related to hardware.
Then try the same settings. Maybe you had less Amps on the brake and battery regen.
No change. Even on a very slight braking from the remote(hence very low braking current), it starts cogging badly before stopping. Not pleasant at all. It’s very similar in unsensored setup too.
I can’t go back to the old setting. I used the mobile app and hence never saved my settings.
BTW, I am on firmware 3.53. Anyone else who’s updated theirs using the version from the app store?
I see that the VESC Tool change log says “Added missing I_max_erpm_fbrake parameter”, I wonder if this has any relation to my problem.
I get the same results as always. Slight vibrations before stand still, which is the default behavior. Tried two different boards with 3.53, VESC 6 and settings suggested by VESC-Tool.
Maybe this is because of the change introduced in fw 3.47
=== FW 3.47 ===
You might be right. I hadn’t updated for a while. So maybe what I am experiencing is from that version.
I’ll check if that’s an option somewhere.
The cogging problem is definitely a bit annoying. If I am forced to ride behind people walking then I have to accelerate and brake mildly and this is where it behaves weirdly. Cogging vibrating and the running smoothly for a bit and doing it again. I am on FOC BTW. If this was a standard behavior, I wouldn’t be saying anything. It happened all of sudden(after updating firmware) and is quite noticeable.
@helpmebuild seems to have the same problem.
Like @taz said could the firmware 3.47 have caused this?
Are you running hall sensors? Weirdly the low speed brake fix can have some undesirable interactions with the low speed braking I have noticed. Try re-running your hall sensor calibration and see if they are nice and evenly spaced. I had one board with a slightly loose hall sensor connection that would go in and out and produce some of the annoying behavior you describe.
Yeah I am running, hall sensors.
I’ll try that.
Edit: @Deodand, everything looks fine and the fact that I can have a very smooth start says that my hall calibration is fine. It’s only the braking that is problematic at low speed.
Each time I connect to vesc tool to adjust my settings (by desktop), it seems that the vesc tool reverts to 12s whenever I change and apply to 10s in the voltage section of motor config. Is this ok?
There is a read motor config button and read app config button if you press those (ones with up arrow and A or M on the right side of vesc tool) you’ll get whatever values are written to the VESC loaded otherwise it just has some defaults in the app too (counter-intuitive but is what it is)
Ah, ok so whatever result comes up after clicking those buttons “is what it is”?
Yah after reading config from the VESC into the vesc tool you should be seeing whatever you set/saved last time. Also if you aren’t using the wizard but setting values in the forms you need to use the write motor config or write app config buttons to save/update the values too.
=== FW 3.54 ===
That is now part of VESC-Tool for your convenience.